Scribd
Upload
66 views19 pages

Session 8 - LTI Systems (III) - Convolution

This document discusses convolution and its properties for linear time-invariant (LTI) systems. Convolution is defined as the integral of one signal shifted over another. The key properties of convolution discussed are commutativity, associativity, distributivity, convolution with the Dirac delta function, linearity, time-invariance, and scaling. Convolution produces the zero-state response of an LTI system and can be used to analyze cascaded and parallel systems. An example calculates the zero-state response of a system given its input and impulse response.

Uploaded by

julian
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
66 views19 pages

Session 8 - LTI Systems (III) - Convolution

This document discusses convolution and its properties for linear time-invariant (LTI) systems. Convolution is defined as the integral of one signal shifted over another. The key properties of convolution discussed are commutativity, associativity, distributivity, convolution with the Dirac delta function, linearity, time-invariance, and scaling. Convolution produces the zero-state response of an LTI system and can be used to analyze cascaded and parallel systems. An example calculates the zero-state response of a system given its input and impulse response.

Uploaded by

julian
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 19

SIGNAL PROCESSING I

LTI Systems:
Convolution
SEBASTIÁN ROLDÁN VASCO
[email protected]
Convolution
The convolution between 𝑥(𝑡) and 𝑔(𝑡) is defined as
follows:

∞
𝑥 𝑡 ∗ 𝑔 𝑡 = න 𝑥 𝜏 𝑔 𝑡 − 𝜏 𝑑𝜏
−∞
Convolution
3 3

2 2
x(t)

g(t)
1 1

0 0

-1 -1
-10 -5 0 5 10 -10 -5 0 5 10
t t

3 3

2 2
x()

g()

1 1

0 0

-1 -1
-10 -5 0 5 10 -10 -5 0 5 10
 
3

2.5
x()
g(-)
Convolution
2
A1
1.5

1
A1>A2>A3
0.5

-0.5
The signal
-1
-10 -8 -6 -4 -2 0

2 4 6 8 10 𝑔(𝑡) shifts

A2 A3
Convolution: Properties
• Commutativa (Proof):
𝑥 𝑡 ∗ 𝑦 𝑡 = 𝑦(𝑡) ∗ 𝑥(𝑡)

• Associative:
𝑥 𝑡 ∗𝑦 𝑡 ∗ 𝑧(𝑡) = 𝑥(𝑡) ∗ 𝑦(𝑡) ∗ 𝑧(𝑡)

• Distributive:
𝑥(𝑡) ∗ 𝑦 𝑡 + 𝑧 𝑡 = 𝑥(𝑡) ∗ 𝑦 𝑡 + 𝑥(𝑡) ∗ 𝑧 𝑡

• Shifting (Proof):
𝑥 𝑡 − 𝑇 ∗ 𝑦 𝑡 = 𝑥 𝑡 ∗ 𝑦(𝑡 − 𝑇)
Convolution: Properties
• Convolution with Dirac delta:

∞
𝑥 𝑡 ∗ 𝛿 𝑡 = න 𝑥 𝜏 𝛿 𝑡 − 𝜏 𝑑𝜏 = 𝑥(𝑡)
−∞
Then
𝛿 𝑡 ∗ 𝛿 𝑡 = 𝛿(𝑡)
Convolution: Properties
• Linearity:
[𝑥1 𝑡 + 𝑥2 𝑡 ] ∗ ℎ 𝑡 = 𝑥1 𝑡 ∗ ℎ 𝑡 + 𝑥2 𝑡 ∗ ℎ 𝑡
• Time invariance:
𝑥 𝑡 − 𝑡0 ∗ ℎ 𝑡 = 𝑦(𝑡 − 𝑡0 )
• Scaling:
1
𝑥 𝛼𝑡 ∗ ℎ 𝛼𝑡 = 𝑦(𝛼𝑡)
𝛼
• Special functions:
𝛿 𝑡 ∗ 𝑥 𝑡 = 𝑥(𝑡)
𝑡
𝑢 𝑡 ∗ 𝑥 𝑡 = න 𝑥 𝜏 𝑑𝜏
−∞
Convolution: Properties
𝑢 𝑡 ∗𝑢 𝑡 =𝑟 𝑡

𝑟𝑒𝑐𝑡 𝑡 ∗ 𝑟𝑒𝑐𝑡 𝑡 = 𝑡𝑟𝑖(𝑡); 𝑟𝑒𝑐𝑡 𝑡 = 𝑢 𝑡 + 0,5 −


𝑢(𝑡 − 0,5)

𝑒 −𝑎𝑡 𝑢 𝑡 ∗ 𝑒 −𝑎𝑡 𝑢 𝑡 = 𝑒 −𝑎𝑡 𝑟(𝑡)


Zero-State response
• Convolution with Dirac delta:

∞
𝑥 𝑡 ∗ 𝛿 𝑡 = න 𝑥 𝜏 𝛿 𝑡 − 𝜏 𝑑𝜏 = 𝑥(𝑡)
−∞
Then
𝛿 𝑡 ∗ 𝛿 𝑡 = 𝛿(𝑡)
Zero-State response
The basic pulse 𝑝(𝑡) is 1 unit height and Δ𝜏 width,
starts at 𝑡 = 0. The signals 𝑥(𝑡) can be represented as
the sum of rectangular pulses.
Zero-State response
𝑥 𝑡 = lim ෍ 𝑥 𝑛Δ𝜏 𝑝(𝑡 − 𝑛Δ𝜏)
Δ𝜏→0
𝜏
𝑝(𝑡 − 𝑛Δ𝜏)
= lim ෍ 𝑥 𝑛Δ𝜏 Δ𝜏
Δ𝜏→0 Δ𝜏
𝜏

𝑥 𝑡 = lim ෍ 𝑥 𝑛Δ𝜏 δ(𝑡 − 𝑛Δ𝜏) Δ𝜏


Δ𝜏→0
𝜏
Zero-State response
Input Output
δ(𝑡) → ℎ(𝑡)
δ(𝑡 − 𝑛Δ𝜏) → ℎ(𝑡 − 𝑛Δ𝜏)
[𝑥 𝑛Δ𝜏 Δ𝜏]δ(𝑡 − 𝑛Δ𝜏) → [𝑥 𝑛Δ𝜏 Δ𝜏]ℎ(𝑡 − 𝑛Δ𝜏)
→
lim ෍ 𝑥 𝑛Δ𝜏 δ(𝑡 − 𝑛Δ𝜏) Δ𝜏 lim ෍ 𝑥 𝑛Δ𝜏 ℎ(𝑡 − 𝑛Δ𝜏) Δ𝜏
Δ𝜏→0 Δ𝜏→0
𝜏 𝜏
𝑥(𝑡) → 𝑦(𝑡)

Hence
∞
𝑦 𝑡 = lim ෍ 𝑥 𝑛Δ𝜏 ℎ(𝑡 − 𝑛Δ𝜏) Δ𝜏 = න 𝑥 𝜏 ℎ 𝑡 − 𝜏 𝑑𝜏 = 𝑥 𝑡 ∗ ℎ(𝑡)
Δ𝜏→0 −∞
𝜏
Zero-State response
• CONCLUSION: The convolution produces the Zero-
State response in a LTI system.

Input 𝑥 𝑡 Impulse response


Output 𝑦 𝑡 = ℎ 𝑡 ∗ 𝑥(𝑡)
ℎ 𝑡
Zero-State response
Cascade systems Equivalent system

𝑥(𝑡) ℎ1 (𝑡) ℎ2 (𝑡) 𝑦(𝑡) 𝑥(𝑡) ℎ1 (𝑡) ∗ ℎ2 (𝑡) 𝑦(𝑡)

𝑦 𝑡 = 𝑥 𝑡 ∗ [ℎ1 (𝑡) ∗ ℎ2 (𝑡)]

Parallel systems Equivalent system


+ ∑
𝑥(𝑡) ℎ1 (𝑡) 𝑦(𝑡)
+
𝑥(𝑡) ℎ1 𝑡 + ℎ2 (𝑡) 𝑦(𝑡)

ℎ2 (𝑡)

𝑦 𝑡 = 𝑥 𝑡 ∗ [ℎ1 𝑡 + ℎ2 (𝑡)]
Zero-State response
The total output of the LTI system can be expressed as
follows:

𝑦 𝑡 = ෍ 𝑐𝑘 𝑒 𝜆𝑘𝑡 + 𝑥 𝑡 ∗ ℎ(𝑡)
𝑘=1

𝑦𝑧𝑖 (𝑡) 𝑦𝑧𝑠 (𝑡)


Zero-Input response Zero-State response
Zero-State response
• Example:
Given 𝑥 𝑡 = 𝑒 −2𝑡 𝑢(𝑡) and ℎ 𝑡 = 𝑒 −𝑡 𝑢(𝑡), find the system’s
Zero-State response.
∞
𝑦 𝑡 = 𝑥 𝑡 ∗ ℎ 𝑡 = න 𝑥 𝜏 ℎ 𝑡 − 𝜏 𝑑𝜏
∞ −∞
= න 𝑒 −2𝜏 𝑢(𝜏)𝑒 − 𝑡−𝜏 𝑢(𝑡 − 𝜏)𝑑𝜏
−∞
Hence:
𝑡
𝑦 𝑡 = න 𝑒 −2𝜏 𝑒 − 𝑡−𝜏 𝑑𝜏 = 𝑒 −𝑡 1 − 𝑒 −𝑡 ; 𝑡 ≥ 0
0
Therefore:
𝑦(𝑡) = 𝑒 −𝑡 1 − 𝑒 −𝑡 𝑢(𝑡)
Zero-State response
Convolution Table:
Zero-State response
Convolution Table:
Zero-State response
Exercises:

You might also like